A behavioural transformer for effective collaboration between a robot and a non-stationary human

TL;DR

This paper introduces BeTrans, a transformer-based meta-learning framework that enables robots to adapt rapidly to non-stationary human behaviors in collaborative tasks, improving human-robot interaction efficiency.

Contribution

We developed BeTrans, a novel transformer-based meta-learning model that enhances robot adaptability to changing human behaviors in collaborative environments.

Findings

BeTrans outperforms state-of-the-art methods in adapting to non-stationary human behaviors.

BeTrans demonstrates rapid adaptation in simulated human-robot collaboration scenarios.

The framework effectively predicts and responds to diverse human behavioral biases.

Abstract

A key challenge in human-robot collaboration is the non-stationarity created by humans due to changes in their behaviour. This alters environmental transitions and hinders human-robot collaboration. We propose a principled meta-learning framework to explore how robots could better predict human behaviour, and thereby deal with issues of non-stationarity. On the basis of this framework, we developed Behaviour-Transform (BeTrans). BeTrans is a conditional transformer that enables a robot agent to adapt quickly to new human agents with non-stationary behaviours, due to its notable performance with sequential data. We trained BeTrans on simulated human agents with different systematic biases in collaborative settings. We used an original customisable environment to show that BeTrans effectively collaborates with simulated human agents and adapts faster to non-stationary simulated human agents than SOTA techniques.